April 21, 2008
I’m calling it, “first date jitters.” Robotic Flower “Flo” has now been assembled, reassembled, plugged in and ready to go. I’m nervous as hell that I’ll screw something up in these last few moments, but if it all goes as smoothly a it has so far, I’ll be practically done by the end of the week. The programming and adjusting the movements will take some time, but the most laborious part seems to be over. I feel like I’m at the end of a good book. I don’t want to finish it. But… if it’s not finished there will have been no point in my having started this project in the first place.
So before I launch into the process of obtaining the Qwerk processor’s unique ID number, followed by the establishment of the wireless connection and synchronization testing of the servos and IR sensors, I think I should jot down a few notes as to what the last week looked like for me and “Flo”, the robot.
First, the middle joint of the stem, which is connected to the H422 servo was much too slack, and the top of the stem kept falling down. I took the two halves apart, leaving the bottom attached to the Qwerk (yay for the tiny screws finally coming in the mail so that I could attach the Qwerk to the stem). I originally fixed that particular joint with a hex-head screw, but at the time I didn’t had a small enough hex-key, so the joint wasn’t tight enough. I had to replace the original ‘L’ bracket with one of my extra brackets to avoid unscrewing and remounting the top of the flower (all servos included). After the reassembly it still seemed to slack, but at least it wasn’t falling down every few seconds. About an hour later I realized that one of the last steps in the assembly involved attaching two small springs to the stem which should hold the whole of the body secure in place, while still permitting the general motion attributed to the servos. Lovely, but of course, I’m still glad I decided to rejoin those sections with a tighter hold because I’m sure it will prevent some unnecessary strain on the robot in the future.
As stated before, I ended up soldering the AC adapter together due to the lag time in the arrival of the coaxial mixed with my general impression that the coaxial was a bit unnecessary. I simply soldered the 12v 5A adapter directly to the open wires of the attached kyosho battery adapter, wrapped them with some electrical tape, and tested out the plug on a small Lego-robotics RCX. Everything worked (woohoo!). Note: I did have to practice on several test wires before I felt comfortable sacrificing my rare kyosho battery parts to my novice skills. Alec K. helped me out by demonstrating on two wires before I attempted to do it myself. But as it turns out, soldering is a little easier than doing fine-glue work on tiny scraps of paper (something I’m very familiar with). The only thing I wasn’t entirely prepared for were the fumes that arise from soldering (mostly when the solder tip and the metal alloy touch). I think I fumigated myself a little, and I had to abandon my work space for several hours to take a nap while the room aired out. Oh well, learn something new every day.
Today I plugged the Qwerk in, tested the power, and obtained the Qwerk ID. After all those butterflies, the 28 digit ID for the robot is now known: 2644036032023432361151761615
What an inspiring string of numbers.
OK, moving on, each one of the servos and the IR sensors have been connected to the extender cables and labeled with an “IR #” or “Servo #” using a small piece of masking tape. Since I like color coding things, I used a different sheet of colored construction paper (backed with poster board) for each of the petals, and labeled each of the wires with either a blue/green marker for the servos, or a brown marker for the IR sensors. When arranging the petals and the wiring the wires had to be gathering and joined to the stem with several zip ties to prevent them from getting in the way. I don’t know if it makes any difference, but I arranged my petals numbering counterclockwise from 1 to 6. Servo # 1 coincides with Servo Port 0. The recipe recommends numbering the servos clockwise from #, but I doubt it makes a difference. I had previously labeled the back of my petals 1-6 before mounting them; removing the labels or remounting would take a few hours.
April 24
Logging in and establishing a connection.
Although I have been able to power on the Qwerk, I’m having some difficulty getting the Qwerk to connect to the internet, or rather, I can’t get it to connect to the RUR software either directly or indirectly. Several attempts at login have failed and I’m a little fuzzy on what’s going wrong. I think I probably need to get into the diagnostics manual to set the configuration for wireless internet, but I don’t know why the Qwerk won’t connect when it’s connected via Ethernet. I’m playing with it right now and hope to get it interacting with the software today.
April 24 (a few hours later)
My attempts to connect directly or through the relay are both failing. I can login successfully to the relay mode, but neither the fakeQwerk or Flo appear online. I have tried connecting the Qwerk directly to the internet and directly to my computer but neither have worked yet. I went into the diagnostics manual tool and reset the default settings. For the version I'm using - 3.x this should mean it automatically connects through relay mode using an IP address determined by DHCP. For version 2 there is a static IP address and subnet mask given as follows:
IP address: 192.168.1.150
Subnet: 255.255.255.0
Once the static IP has been set on the Qwerk, and the Qwerk has been connected to the computer or ethernet and allowed to boot up, I should be able to navigate to this website which allows me to change settings on my qwerk without using the flashing LEDs on the board itself.
http://192.168.1.144
I went into my control panel but couldn't quite find where to change my TCP/IP settings to those listed above. I did however change the settings that allow virtual private connections figuring that a direct connection would have to be enabled for my two machines to communicate.
At this point my best guess is that my lack of connectivity could be due to one (or a combination) of the following problems:
1) My TCP/IP settings need to be adjusted with the statsic IP address given
2) A wireless adapter is needed for the Qwerrk (this is actually a bug that has been found with the 3.x Qwerks... I don't get it)
3) I have done something wrong in the wiring stages, or in connecting the Qwerk to the computer/Ethernet etc.
4) I am missing something in the login phase with connecting to the sofware.
I'll keep troubleshooting for a while, but I might need a hand with this part.
1.) Java console documentation during login attempts 2/24
2.) Images of assembly progress
Brief Visual journey through the assembly process
For now, just take a look at the attached image files to see what the things looks like. I still need to put them in order and given them little captions, but the file names give a fairly straightforward description.
![[paper clip]](/courses/source/wiki_images/paper_clip_tilt.png)
name last modified size
